The present invention relates to a process for producing a nanocomposite material which is formed from    a) at least one inorganic or organometallic phase; and    b) an organic polymer phase.
Nanocomposite materials, i.e. polymer-based composite materials which comprise, in an organic polymer phase, an inorganic phase, for example an inorganic (semi)metal oxide phase with dimensions below 500 nm, especially below 100 nm (hereinafter, also nanoscale phase or, in the case of a particulate phase, nanoscale particles), owing to the large interface between the nanoscale inorganic phase and the organic polymer phase, possess a high potential with regard to their chemical, physical and mechanical properties, which cannot be achieved by milli- or microscale dispersions of conventional inorganic constituents in polymer phases (R. P. Singh, et al., J. Mater. Sci. 2002, 37, 781).
The processes known to date for producing inorganic-organic nanocomposites are based on direct mixing of nanoparticles or exfoliated sheet silicates with a polymer in solution or the melt, the in situ preparation of the organic phase by polymerizing organic monomers in the presence of inorganic nanoparticles or exfoliated sheet silicates, sol-gel techniques and combinations of these measures (see, for example, for incorporation of nanoparticles into a polymer melt: Garcia, M.; et al., Polymers for Advanced Technologies 2004, 15, 164; for polymerization of organic monomers in the presence of inorganic nanoparticles or exfoliated sheet silicates see: M. C. Kuo et al., Materials Chemistry and Physics 2005, 90(1), 185; A. Maity et al., Journal of Applied Polymer Science 2004, 94(2), 803; Y. Liao et al. (Polymer International 2001, 50(2), 207; and WO 2001/012678; for preparation of an oxide phase by hydrolysis of oligomeric alkoxysiloxanes in a polymer solution or melt see WO 2004/058859 and WO 2007/028563).
The established prior art methods are associated with a series of disadvantages. Firstly, many of them remain restricted to composites of organic polymers which are either soluble in organic solvents or melt without decomposition. In addition, it is generally possible in this way only to introduce small amounts of inorganic phase into the nanocomposite material. Owing to the usually high agglomeration of the nanoparticles and the enormously high shear forces which are necessary as a result, fine distribution of the nanoparticles in a relatively large amount is barely possible. A great disadvantage of nanocomposite production by in situ preparation of the organic polymer phase in the presence of nanoparticles is the formation of nanoparticle agglomerates which occurs, such that inhomogeneous products form. As a result, the advantage of the nanoparticles, that of forming extensive interfaces with the polymer as a result of their large surface area, cannot be utilized. In the case of use of pulverulent nanofillers, owing to the small particle size, there is additionally a high risk to health during compounding owing to the dust formation which occurs and the ability of the nanoparticles to reach the lungs. The in situ preparation of the inorganic phase by a sol-gel process in a polymer melt or solution generally leads to poorly reproducible results or requires complex measures to control the hydrolysis conditions.
Spange et al., Angew. Chem. Int. Ed., 46 (2007) 628-632 describe a novel route to nanocomposite materials by cationic polymerization of tetrafurfuryloxysilane TFOS and difurfuryloxydimethylsilane according to the following schemes:

Spange et al. propose the term “twin polymerization” for this kind of polymerization.
The polymerization of TFOS under cationic conditions forms a composite material which has a silicon dioxide phase and an organic polymer phase composed of polyfurfuryl alcohol PFA. The dimensions of the phase domains in the composite material thus obtained are in the region of a few nanometers. In addition, the phase domains of the silicon dioxide phase and the phase domains of the PFA phase have a co-continuous arrangement, i.e. both the PFA phase and the SiO2 phase penetrate one another and essentially do not form any discontinuous regions. The distances between adjacent phase interfaces or the distances between the domains of adjacent identical phases are extremely small and are on average not more than 10 nm. There is no macroscopically visible separation in discontinuous domains of the particular phase.
It is assumed that the specific phase arrangement and the small distances between adjacent phases are a consequence firstly of the kinetic coupling of the polymerization of the furfuryl units, and secondly of the formation of the silicon dioxide. Consequently, the phase constituents form more or less synchronously and there is a phase separation into the inorganic phase and the organic phase as early as during the polymerization of TFOS.
In the course of polymerization of DFOS, in contrast, no formation of a composite material is observed. Instead, PFA and oligomeric dimethylsiloxane are formed, the latter separating out as an oil.
The twin polymerization described by Spange et al. solves a series of problems in the production of nanocomposite materials and especially avoids the use of nanomaterials. However, the nanocomposite materials described by Spange et al. are unsatisfactory in various aspects and have only moderate strength and are comparatively brittle.